Type 2 diabetes (T2D) shows complex inheritance, indicating a causal role for multiple inherited DNA variants. Genome wide association studies (GWAS) have now mapped over 20 novel loci where common variants are associated with risk of T2D. Despite this progress, identified risk alleles explain relatively little of the overall variation in T2D risk. To fully understand the genetic architecture of T2D we need to move from locus to gene to pinpoint specific causal gene(s) responsible for observed associations. We need to address allelic heteroaeneitv. where T2D genes are likely to have multiple different common and rare mutations. We need to explore ethnic variation, where the specific complement of gene mutations contributing to T2D are likely to vary in frequency and effect size across ethnic groups. We hypothesize that: (1) each region identified by GWAS contains at least one causal T2D gene, influenced by at least one common functional variant; (2) in addition to the index variant identified by GWAS, one or more additional common variants in each locus influence T2D; (3) in addition to common variants, each gene may harbor rare mutations that more strongly influence risk of T2D, and (4) the identities, frequencies and effects of these variants vary across multiple ethnic groups representative of the US population. To address these hypotheses we propose three Specific Aims. (1) Bring together multiethnic samples representative of the US population, drawn from the Jackson Heart Study, Framingham Heart Study, Multi-Ethnic Cohort Study, and Diabetes Prevention Program, that together include -29,000 individuals with T2D phenotypes and DNA; (2) Identify and fine-map common variants at each locus in each ethnic group by leveraging our multi-ethnic design and emerging data from the 1000 Genomes Project; and (3) Identify rare causal mutations at each locus by performing deep sequencing of all coding exons in each ethnic group. The co-investigators have extensive experience in complex disease genetics and genomics, next-generation sequencing, statistical genetics, metabolic physiology and epidemiology, and have a long track-record of effective collaboration and leadership that, combined with a large multiethnic, well-phenotyped sample, we hope can contribute to RFA-DK-09-004. RELEVANCE (See instructions): Genetic studies of type 2 diabetes (T2D) have identified new genomic risk regions. We will look in these regions for genes, define variation within genes and variation in different people by bringing together -29,000 individuals from ethnic groups representing the US population, map genes in each region, and identify mutations by detailed DNA analysis, leading to better prevention and treatment of T2D.